interpretability

A roughly five-hour Lex Fridman podcast episode with Anthropic CEO Dario Amodei, plus Amanda Askell on Claude's character and Chris Olah on mechanistic interpretability. It covers Claude, AGI and the future of AI.

A short demo of the Deep Visualization Toolbox, which shows the activations inside a convolutional neural network as it processes live camera input. Viewers see individual units responding to edges, faces and text.